Spoken language acquisition in children with
prelingual hearing loss: A review of their
morphosyntactic abilities at the sentence level
Analí R. TABOH 1,2,3, Diego E. SHALOM 1,3, Carolina A. GATTEI 1,3,4
1Departamento de Física, FCEyN, Universidad de Buenos Aires and IFIBA, CONICET2Facultad de Filosofía y Letras, Universidad de Buenos Aires3Laboratorio de Neurociencia, Universidad Torcuato di Tella4Pontificia Universidad Católica Argentina
Version date: August 23, 2021
Contact:
[email protected] (Analí Taboh)
Abstract
The language development of children with hearing impairment who access spoken language
through hearing devices (CHI) usually lags behind that of children with typical development.
In order to improve therapeutic and educational practices for these children, it is important
to understand which specific aspects of language represent difficulties for them. This work
reviews the evidence on CHI’s morphosyntactic abilities at the sentence level by focusing on
their comprehension and production of specific types of structures: simple sentences, topi-
calized sentences, passives, sentences with relative clauses, and wh-questions. It also reviews
demographic, clinical, and cognitive factors that introduce variability in CHI’s performance, as
well as different accounts of their difficulties at the sentence level.
Keywords: children with hearing impairment, sentence comprehension and production,
complex sentences, morphosyntax.
Introduction
Prelingual hearing loss (i.e. hearing loss with onset previous or simultaneous to language ac-
quisition, typically up to age three; HL) has an impact not only on the acquisition of spoken
language but also on academic, emotional, and social development (Dammeyer, 2010; Qi and
Mitchell, 2012). In the long term, it affects people’s opportunities for employment and social in-
tegration. Even though assistive devices such as hearing aids (HAs) and cochlear implants (CIs)
improve hearing considerably, children with hearing impairment acquiring spoken language
(CHI) usually do not achieve the same level of language development as children with typical
development (CTD). Several studies have reported this for children with different degrees of
HL and who speak different languages (Tomblin, Barker, Spencer, Zhang, and Gantz, 2005; De-
lage, 2008; Geers, Moog, Biedenstein, Brenner, and Hayes, 2009; Inscoe, Odell, Archbold, and
Nikolopoulos, 2009; Niparko et al., 2010; Caselli, Rinaldi, Varuzza, Giuliani, and Burdod, 2012;
Nittrouer, Sansom, Low, Rice, and Caldwell-Tarr, 2014; Soleymani, Mahmoodabadi, and Nouri,
2016). It is crucial to research which aspects of language are affected when a child with HL ac-
quires spoken language, the reasons behind these limitations, and which factors influence the
process. This research will then allow to improve and fine-tune therapeutic and educational
practices so that children with HL can have better language attainment and, thus, the same
opportunities as their peers with normal hearing.
In particular, the aim of this work is to provide a comprehensive review of what has been
reported regarding the morphosyntactic abilities of CHI at the sentence level. Rather than
providing an overall view, we wish to pinpoint which specific aspects of morphosyntax present
difficulties for CHI. The review is organized as follows: We first review the evidence on CHI’s
comprehension and production of sentences with different structures. Next we describe pos-
sible factors that introduce variability in CHI’s performance. We then review the evidence on
how different demographic, clinical, and cognitive factors are related to CHI’s performance. Fi-
nally, we present and discuss possible accounts of CHI’s difficulties in sentence comprehension
and production.
In this review, we have included those publications that meet the following criteria: (i) they
include CHI whose first language is an oral language; (ii) they evaluate the acquisition, compre-
hension and/or production of sentences in spoken language in CHI; (iii) they present separate
results for different syntactic structures. Because we focus on CHI’s morphosyntactic abilities
1
in spoken language, and given that, as mentioned above, access to sound is much improved
by hearing devices, we have excluded from this review those studies which do not indicate
whether the CHI used hearing devices. See Table A1 in Appendix A for the list of included and
excluded studies.
Studies targeting this population vary widely in their experimental designs. One way in
which they differ is in the use of standardized tests, experiments designed to evaluate specific
structures, or speech samples. At the sentence level, there is vast evidence that CHI perform
below CTD on standardized tests (Young and Killen, 2002; Geers, Nicholas, and Sedey, 2003;
Nikolopoulos, Dyar, Archbold, and O’Donoghue, 2004; Schorr, Roth, and Fox, 2008; Geers
et al., 2009; Boons et al., 2013; Mandal, Kumar, and Roy, 2016; Soleymani et al., 2016, but
see Briscoe, Bishop, and Norbury, 2001; Norbury, Bishop, and Briscoe, 2001; Hansson, Fors-
berg, Löfqvist, Mäki-Torkko, and Sahlén, 2004; Halliday and Bishop, 2005; Duchesne, Sutton,
and Bergeron, 2009; Halliday, Tuomainen, and Rosen, 2017). Indeed, standardized tests like
the Test for Auditory Comprehension of Language (Woolfolk-Carrow, 1999), the Reynell Devel-
opmental Language Scale (Reynell & Gruber, 1990), the Clinical Evaluation of Language Funda-
mentals (Semel,Wiig, and Secord, 1995; Semel-Mintz,Wiig, and Secord, 2003), and the Test for
Reception of Grammar (Bishop, 2003) have been commonly used for evaluating morphosyn-
tactic performance in these children. The main limitation of standardized tests like these is
that they assess a range of structures but do not analyze each of them separately. As a conse-
quence, while they provide a general picture of morphosyntactic abilities, they do not allow to
identify which specific structures present difficulties for CHI. For this reason, in this work we
will not include studies using this type of assessment; instead, we will include studies using
experiments or speech samples to evaluate specific structures. We will focus on the following:
simple sentences, topicalized sentences, passives, sentences with relative clauses, and open
questions or wh-questions (see examples of each structure in Table 1).
Another way in which studies targeting CHI differ is how experimental groups are formed.
The performance of CHI is compared to that of CTD matched (either at the group or the in-
dividual level) on chronological age (Boons et al., 2013; Halliday et al., 2017), hearing age (i.e.
time since an adequate hearing device has been received; Caselli et al., 2012; Schouwenaars,
Finke, Hendriks, and Ruigendijk, 2019), language level (typically measured in mean length of
utterance or vocabulary; Volpato, 2010; Volpato and Vernice, 2014), school grade (Nittrouer et
al., 2014) or nonverbal cognitive ability (Nittrouer et al., 2014). Also, groups of CHI vary in how
2
Table 1: Structures this review focuses on and examples of each
Structure Example
Simple active sentences The dog is chasing the cat.
Topicalized sentences The cat, the dog is chasing.
Passives sentences The cat is being chased by the dog.
Sentences with relative clausesThe dog that is chasing the cat is white.
The cat that the dog is chasing is white.
Who is chasing the cat?
Who is the dog chasing?
Which dog is chasing the cat?Wh-questions
Which cat is the dog chasing?
homogeneous or heterogeneous they are. For example, while the chronological age range of
the CHI group in Caselli et al. (2012) is 3;8 to 5;5, Friedmann and Haddad-Hanna (2014) include
CHI aged 9-21 years; whereas the CHI in Geers et al. (2003) range in hearing age from 4 to
6 years, the CHI in DeLuca (2015) range from 4 to 10. Also, while Ruigendijk and Friedmann
(2017) test CHI with moderate to profound HL, Penke andWimmer (2018) test CHI with mild to
severe HL, and Volpato and Vernice (2014) focus on CHI with profound HL; and whereas Lee,
Sung, and Sim (2018) report on CHI with CIs, Werfel, Reynolds, Hudgins, Castaldo, and Lund
(2021) include some CHI with HAs and others with CIs. Careful examination of these variables
is crucial for accounting for linguistic performance in this population. In the following sections
we review the evidence presented by those studies that have focused on how CHI with differ-
ent degrees of HL and who use different languages comprehend and produce specific types of
syntactic structures.
Simple sentences
Studies that include simple active sentences with canonical word order as part of their stimuli
have generally found that groups of CHI across the range of HL performat ceiling on these struc-
tures, both in comprehension (mild to moderate HL: Norbury, Bishop, and Briscoe, 2002 for
English; moderate to severe HL: Friedmann, Novogrodsky, Szterman, and Preminger, 2008 for
3
Hebrew; moderate to profound HL: Friedmann and Szterman, 2006; Friedmann, Rizzi, and Bel-
letti, 2016 for Hebrew; Friedmann and Haddad-Hanna, 2014 for Palestinian Arabic; Ruigendijk
and Friedmann, 2017 for German; degree of HL not stated: Fujiyoshi et al., 2012 for Japanese;
Lee et al., 2018 for Korean) and in production (mild to moderate HL: Delage, 2008 for French;
moderate to severe HL: Friedmann and Szterman, 2011 for Hebrew; moderate to profound HL:
Ruigendijk and Friedmann, 2017 for German). Nevertheless, some studies have found differ-
ences between CHI and CTD in performance in these structures. Ruigendijk and Friedmann
(2017) found that German-speaking CHI with moderate to profound HL aged 9-13 years re-
peated simple SVO sentences with over 90% accuracy (as a group), but still significantly below
a group of age-matched CTD; the authors attributed this difference to the CTD performing at
100% accuracy. Also in repetition, Caselli et al. (2012) reported that Italian-speaking CHI with
profound HL aged 3-5 years were significantly less accurate than CTD matched on chronologi-
cal age, but not when compared to younger CTDmatched on hearing age1. Novogrodsky, Meir,
andMichael, 2018 analyzed the repetition of short simple sentences with nominal, unergative,
SV unaccusative, and VS unaccusative structure in Hebrew-speaking CHI with mild to profound
HL aged 2-3 years and age-matched CTD, and they found amarginal interaction between group
and sentence structure. Regarding the age of acquisition of simple sentences, Fujiyoshi et al.
(2012) reported that Japanese-speaking CHI reached levels of comprehension above 80% at
age 6 and above 90% at age 8, while CTD were already over 90% accurate at least as young
as 5 years (the study did not include children younger than 5 years). Analyses of individual
performance also indicate that CHI’s comprehension and production of simple sentences are
not always entirely preserved. Across different experiments, Ruigendijk and Friedmann (2017)
showed that 10-20% of the CHI performed significantly below the control groups.
While simple active sentenceswith canonicalword order are usually preserved in CHI, there
is extensive evidence that more complex sentences pose considerable difficulties for this pop-
ulation. Below, we summarize the findings on several complex structures and briefly discuss
possible lines of future investigation.1However, it should be noted that, apart from simple SVO sentences, the repetition task in this study included
SV sentences with unergative verbs and SV sentences where the verb was transitive but the object was omitted
(though still predictable, e.g. with verbs like ‘iron’).
4
Topicalized sentences
The acquisition of sentences with topicalized object has been studied in CHI in comprehen-
sion and repetition. Several works have examined two different topicalization structures in
Hebrew: OSV structures, which only differ from the canonical SVO order in that the object is
in first position, and preserve the SV order, and OVS, where the position of each constituent
relative to the others is different than in the canonical order (Friedmann and Szterman, 2006;
Friedmann et al., 2008; Friedmann and Szterman, 2011; Friedmann et al., 2016). Friedmann
and Szterman (2006) found that CHI with moderate to profound HL comprehended topical-
ized OSV structures similarly to CTD approximately 2.5 years younger – and at ceiling – but
topicalized OVS structures significantly worse, with 9 of the 20 CHI performing at chance. Sim-
ilarly, Friedmann et al. (2008) found that CHI with moderate to severe HL comprehended OVS
topicalization structures significantly below younger CTD. Accuracy in the comprehension of
this type of sentences ranged from 59% (in Friedmann et al., 2008) to less than 70% (in Fried-
mann and Szterman, 2006). It would seem from these results that OSV topicalization is easier
for Hebrew-speaking CHI than OVS topicalization. However, Friedmann and Szterman (2006)
suggested that CHI’s ceiling performance on OSV sentences might not have always been the
result of a correct syntactic representation, because when hearing these sentences they often
repeated the SV part and then pointed, within the correct picture, to the agent of the action,
not to the theme. As for OVS structures, Friedmann et al. (2016) argued that the fact that
CHI did not present systematic below-chance performance (only 2 out of 18 CHI in their study
performed below-chance) indicates that they did not interpret these sentences as SVO, which
suggests that they were sensitive to the case information provided by the object marker, but
could not use it to build the correct interpretation.
For German, Ruigendijk and Friedmann (2017) (experiment 2) reported that CHI with mod-
erate to profound HL showed comprehension of OVS topicalized sentences at chance level,
and significantly worse than CTD aged as the youngest CHI. Fujiyoshi et al. (2012) compared
the age of acquisition of OSV structures (which they refer to as ‘reversed word order object-
agent-verb, OAV’) in Japanese-speaking CHI and CTD, with a sentence-picturematching task. A
child was considered to have acquired a given structure when she selected the correct picture
for at least three of the four sentences presented for that structure. The authors found that
OSV structures were acquired later by CHI (8% of 5-year-olds, 23.1% of 6-year-olds, and 47.1% of
5
8-year-olds) than by CTD (52.3% of 5-year-olds, 67.9% of 6-year-olds, and 82.1% of 8-year-olds).
In repetition tasks, to the best of our knowledge only OSV structures have been tested.
Hebrew-speaking CHIwithmoderate to severeHL repeated these structureswith 61%accuracy,
over 30% less than the younger CTD (Friedmann & Szterman, 2011). While there is convergent
evidence that sentences with topicalized object are challenging for CHI, further investigation is
needed to clarify whether this is true only for those with OVS order or also for sentences with
OSV order.
Passive sentences
Most studies on the comprehension and production of passive sentences by CHI have provided
evidence of good performance, even if differences with CTD have been observed. Norbury et
al. (2002) found that English-speaking CHI with mild to moderate HL performed similarly to
age-matched CTD in a sentence-picture matching task including full passives, short progres-
sive passives, and short ambiguous passives. Ruigendijk and Friedmann (2017) found similar
results, also with a sentence-picture matching task, for German-speaking CHI with moderate
to profound HL. In a repetition task, these children showed group performance above 90% but
still significantly below age-matched CTD, which the authors attributed to the ceiling perfor-
mance of CTD. It should be noted, however, that the childrenwho participated in the repetition
tasks in Ruigendijk and Friedmann, 2017 had passed a simple sentence repetition screening
test. Only those who made errors in no more than one sentence, including omission or substi-
tution of determiners (which in German are marked by case), were included. Given that there
is evidence that German-speaking CHI omit and substitute determiners more often than CTD
matched on mean length of utterance (Szagun, 2004), the CHI in Ruigendijk and Friedmann
(2017) might not be representative of the population.
Lee et al. (2018) reported that Korean-speaking CHI with CIs comprehended passive sen-
tences with 16% less accuracy than age-matched CTD. The authors also reported that, in a
logistic regression in which group was entered as the dependent variable and active and pas-
sive sentence scores were entered as independent variables, passive (but not active) sentence
scores were significantly associated with the identification of CHI among all children included
in the study, i.e., children with lower passive sentence scores were more likely to be classified
as CHI. This evidence suggests that CHI comprehend and produce passive sentences relatively
6
well. There are, however, some cases of poor performance among the CHI included in these
studies: 2 out of the 19 CHI in Ruigendijk and Friedmann (2017) performed at or below chance
in the sentence-picturematching task, and accuracy in Lee et al. (2018) ranged from 10 to 100%
(while accuracy in CTD ranged from 70 to 100%).
Nevertheless, one study found a considerable delay in CHI’s acquisition of passive sen-
tences. Fujiyoshi et al. (2012) found that Japanese-speaking CHI acquired these structures
much later than CTD. The CTD who comprehended correctly at least three of the four sen-
tences in passive voice included in the testwere 22.7%of 5-year-olds, 64.3%of 6-year-olds, and
75.0% of 8-year-olds, whereas in the CHI group they were 8.0% of 5-year-olds, 0% of 6-year-
olds, 42.9% of 7-year-olds, 35.3% of 8-year-olds, 53.8% of 9-year-olds, 41.7% of 10-year-olds,
63.6% of 11-year-olds, and 75.0% of 12-year-olds. It is interesting to note that passive sentences
were the last structure to be acquired by CTD, but not by CHI, who acquired sentences with
relative clauses last.
As can be seen, the number of studies that have examined the comprehension and produc-
tion of passive sentences in groups of CHIwith access tomodern hearing devices and treatment
are very scarce. In many languages, including English and German, the passive voice is formed
with an auxiliary verb and presents the agent with a prepositional phrase, all of which might
make the structure more easily identifiable than other types of sentences. More studies in
different languages and with CHI with varying degrees of HL are needed to elucidate whether
these structures are problematic. In particular, future research should look into languages
with alternative markers of passive voice. Analyzing whether the performance with passive
sentences is related to how the passive construction is formed in each language might provide
insight into CHI’s language development.
Sentences with relative clauses
Davis and Blasdell (1975) examined the comprehension of subject relative clauses in English-
speaking CHI with mild to moderate HL aged 6-9 years. Their sentence-picture matching task
included four pictures that represented different simple sentences: the main sentence (the
correct answer, which likely showed comprehension of both clauses), the embedded clause,
the object of the relative clause performing a new action to the object of the main clause,
and the relative clause with thematic role reversal. While both the CHI and age-matched CTD
7
selected the correct picture more often than the other three, the CHI selected it significantly
less often than the CTD, and the rate of correct answers was not influenced by age or degree
of HL. The CHI also differed from the controls in their pattern of errors: they chose the two
pictures that did not represent any part of the orally-presented sentence more often.
Volpato and colleagues (Volpato, 2010, 2020; Volpato & Vernice, 2014) have examined the
comprehension and production of relative clauses in Italian-speaking CHI with profound HL.
They found the same gradient of difficulty present in typical development (i.e., subject relatives
are easier than object relatives and object relatives with preverbal embedded subject are eas-
ier than object relatives with postverbal embedded subject). While the CHI performed overall
significantly below CTD in both comprehension and production, the patterns for each type of
relative clause differed in the two tasks. CHI aged 7-10 years comprehended subject relatives
and a subtype of object relatives with preverbal subject similarly to younger CTD matched on
morphosyntactic abilities (Volpato, 2010) and to CTDmatched on age (Volpato, 2020), but they
comprehended another subtype of object relatives with preverbal subject and object relatives
with postverbal subject significantly worse than both CTD groups. These studies also looked
into how the pattern of comprehension of object relatives with preverbal subject differed be-
tween CTD and CHI according to whether the number features between two noun phrases
matched or not. CHI were significantly less accurate than younger and same-age CTD in sen-
tences with this type of relatives in the mismatch condition, but not different from either CTD
group in the match condition. Additionally, the CHI showed a different gradient of difficulty
among number match and mismatch conditions than both CTD groups: while the CTD were
more accurate in the mismatch condition, the CHI were more accurate in the match condition.
The author proposed that CHI are less sensitive to number features and it is therefore more
difficult for them to exploit the number cues on the verb in order to comprehend the sentence.
Volpato and Vernice (2014) studied relative clause production with an elicitation task in
which children were required to express which of two characters they preferred. They found
that CHI produced fewer correct subject andobject relative clauses than a groupof age-matched
CTD, a group of younger CTD matched on morphosyntactic abilities, and another group of
younger CTD matched on hearing age. The CHI also differed from the CTD on the types of sen-
tences they produced. Whereas Italian-speaking CTD usually produce more object relatives at
first and then shift to passive relatives at later stages of their development, CHI as a groupwere
at an intermediate position between the younger and age-matched CTD, producing fewer ob-
8
ject relatives and more passive relatives than the former, and more object relatives and less
passive relatives than the latter. At the individual level, however, some CHI performed similarly
to same-age CTD, others similarly to younger CTD, and yet others adopted strategies which are
attested in younger CTD but never or rarely in CTD of their age.
In Hebrew- and Palestinian Arabic-speaking CHI, Friedmann and colleagues (Friedmann
and Szterman, 2006; Friedmann et al., 2008; Friedmann and Haddad-Hanna, 2014) found the
same gradient of difficulty in relative clause comprehension as Volpato and colleagues found
in Italian-speaking children: subject relatives were easier than object relatives and (in the case
of Palestinian Arabic) object relatives with preverbal subject were easier than object relatives
with postverbal subject. When compared to younger CTD who performed at or near ceiling
in all types of relatives, both groups of CHI had impaired comprehension of object relatives.
However, Friedmann et al. (2008) found that the performance ofHebrew-speaking CHI differed
considerably between object relatives with a gap and object relatives with a resumptive pro-
noun (which is optional in the object position in Hebrew): an additional test including 8 of the
CHI in their study revealed that the comprehension of all but one of themwas between 95 and
100% accurate for object relatives with a resumptive pronoun. Contrary to Hebrew, in Pales-
tinian Arabic the presence of the resumptive pronoun, which is obligatory in object relatives,
did not facilitate comprehension. Friedmann and Costa (2011) proposed that this difference
can be explained by the different status of the resumptive pronoun in each language: whereas
in Hebrew it is a strong pronoun, and so can be assumed to be inserted post-syntactically,
in Palestinian Arabic it is a clitic, and thus the result of a derivation, so CHI cannot rely on it
to interpret wh-movement. Apart from the difficulty with object relatives, common to both
Hebrew- and Palestinian Arabic-speaking CHI, the latter were significantly more impaired in
object relatives with postverbal subject than with preverbal subject.
Interestingly, while both groups showed relatively good comprehension of subject rela-
tives, the Hebrew-speaking CHI (Friedmann & Szterman, 2006; Friedmann et al., 2008) per-
formed no differently from the younger CTD controls, but the Palestinian Arabic-speaking CHI
were significantly below in Friedmann and Haddad-Hanna (2014), even though in all these
studies the CTD groups were on average 2 or more years younger. The Hebrew-speaking CHI
in Friedmann et al. (2008), however, did perform significantly below a group of age-matched
CTD.
9
Friedmann and colleagues (Friedmann & Szterman, 2006; Friedmann et al., 2008) also ex-
amined the production of relative clauses in Hebrew-speaking CHI with moderate to profound
HL. In two elicitation tasks using picture preference and picture description, the CHI had diffi-
culties producing both subject and object relatives, but did much worse on the latter. In many
cases, instead of producing object relatives, they produced subject relatives, sentences with-
out relative clauses or ungrammatical sentences. Whenever they produced object relatives,
they tended to include a resumptive pronoun in object position, which is a way of avoiding
movement, since the resumptive pronoun fills the position that would correspond to the gap.
The authors pointed out that, while object relatives with resumptive pronouns are grammati-
cal in Hebrew, they are characteristic of much younger children than the CHI included in their
study. The CHI sometimes (though much less often) produced resumptive pronouns in sub-
ject relatives too, which are ungrammatical in Hebrew and were almost never produced by
the CTD. They also doubled the noun phrase in both subject and object relatives, which is also
ungrammatical and never produced by CTD.
Ruigendijk and Friedmann (2017) reported that German-speaking CHI with moderate to
profound HL comprehended subject and object relatives significantly worse than CTD aged as
the younger CHI and showed greater difficulties in object relatives, in which they performed at
chance level. These authors also examined the production of subject relatives with a repetition
task and found a significant difference between groups for center-embedded subject relatives
but not for right-branching ones.
Delage (2008) (also reported in Tuller and Delage, 2014) studied the morphosyntactic abil-
ities of French-speaking CHI with mild to moderate HL employing different methodologies,
including a standardized battery, elicitation tasks and speech samples. The analysis of spon-
taneous speech samples taken approximately two years apart from each other showed that,
while at age 6-11 years old CHI produced significantly fewer relative clauses than age-matched
CTD, this was not the case at age 7-13. In a task that elicited the production of subject and
object relative clauses, Delage found that CHI aged 7-11 produced significantly fewer subject
relatives than a group of 6-year-old CTD and than CHI aged 11-13, and that the latter (but not
the youngest CHI) produced significantly fewer object relatives than the 6-year-old CTD. How-
ever, the author pointed out that the low performance of the youngest CHI on subject relatives
was probably due to the very poor performance of three children in that group. As regards ob-
ject relatives, there was a general tendency to produce causative passives or passive subject
10
relatives instead, which are also appropriate responses to the task. This tendency was high in
6-year-old CTD and higher in the eldest CHI (75 and 92% of non-object relative clauses, respec-
tively). This task, thus, did not allow to draw conclusions about CHI’s ability to produce object
relatives.
Werfel et al. (2021) also examined the production of relatives clauses in the spontaneous
speech of CHI. Their analyses included subject, object, oblique, adjunct, and headless relatives.
The speech samples of a group of English-speaking 4-year-old CHI was compared to those of a
group of same-age CTD and a group of CTD matched on mean length of utterance measured
in morphemes. While the CHI did not differ significantly from either CTD group in the number
of correct relative clauses produced for any of the five types of relatives analyzed, they did
produce a lower percent of correct subject relatives than both CTD groups. According to the
authors, this measure indicated that, when attempting to produce subject relatives, the CHI
made errors more often than the CTD. It is interesting that this was the case for subject but
not for object relatives, given the usual finding, mentioned above, that object relatives present
more difficulties than subject relatives.
As regards the age of acquisition, Fujiyoshi et al. (2012) found that Japanese-speaking CHI
acquired sentences with both subject and object relative clauses much later than CTD. Over
70%of the 6-year-old CTD and over 90%of the 8-year-old CTD comprehended correctly at least
three of the four subject relatives included in the test. In contrast, only around 30% of the 8-
year-old CHI and 73% of the 11-year-old CHI met this criterion. In the case of object relatives,
more than 80% of the 6-year-old CTD and almost 90% of the 8-year-old CTD comprehended
at least three out of four sentences correctly, whilst only around 50% of the 8-, 9-, 10-, and
11-year-old CHI met the criterion.
As can be gathered from this section, sentences with relative clauses have received a lot of
attention from researchers studying the spoken language abilities of CHI. Almost all studies re-
port impaired comprehension and production of object relatives in CHI, even when compared
to younger CTD. In languages that allow both preverbal and postverbal object relatives, the lat-
ter are more affected. As for subject relatives, the evidence is less straight-forward, with some
studies reporting that comprehension and production of these structures is relatively spared,
and others that it is worse than that of younger CTD. Furthermore, some studies showed that
the error patterns of CHI in tasks of both comprehension and production of relative clauses
11
differ from those observed in CTD.
Wh-questions
Wh-questions have also received considerable attention in the study of language abilities of
CHI. Wh-question comprehension has been most frequently assessed with point-in-picture
tasks, in which the participant is shown a picture of three characters interacting, where two
are similar and one is different, e.g. a mother combing a girl on her left and being combed by
another girl on her right, a boy being sprayed by an elephant on his left and spraying another
elephant on his right. The participant is then asked a question, such asWhich elephant is the
boy spraying?, and has to answer by pointing at one of the characters in the picture.
Friedmann and colleagues used this task to study question comprehension in Hebrew- and
Palestinian Arabic-speaking CHI. Friedmann and Szterman (2011) reported that a small group of
Hebrew-speaking CHI withmoderate to severe HL aged 9-12 comprehended subject and object
who-questions similarly to a group of CTD on average 2 years younger, but they comprehended
subject and object which-questions significantly worse. Among these four types of questions,
the most difficult ones for CHI were object which-questions, on which all 11 CHI in the sample
performed below the CTD group. A similar pattern of results for the comprehension of subject
and object which-questions was reported in Friedmann and Haddad-Hanna (2014) for Pales-
tinian Arabic-speaking CHI and adolescents with moderate to profound HL aged 9-21 (mean
age 13;7). This group of CHI performed significantly below a group of much younger CTD aged
6-8 on subject and object questions, the latter both with preverbal and postverbal subject.
They also displayed the same gradient of difficulty found for sentences with relative clauses
and topicalized sentences: object questions were comprehended more poorly than subject
questions, and within object questions those with postverbal subject were harder than those
with preverbal subject.
Friedmannand Szterman (2011) also investigatedwh-question production inHebrew-speaking
CHI with moderate to severe HL aged 9-12. In one experiment they compared their elicited
production of who-questions with that of CTD two years younger. In this elicitation task, the
children saw a picture in which one character was performing an action on another but one
of them was concealed; the children had to ask a question for which the answer would be the
concealed character. In contrast to what the authors found for who-question comprehension,
12
the CHI produced significantly fewer subject and object who-questions than the CTD on aver-
age 2 years younger. Their production of subject questions was, however, significantly better
than their production of object questions. In another experiment, Friedmann and Szterman
compared the repetition of which-questions in the same group of CHI and a group of 5-year-
old CTD. They found the same pattern of results previously found for the comprehension of
this type of questions and for the elicited production of who-questions, even though the age
difference with the CTD group who participated in the repetition task was much larger.
Two studies investigated the comprehension of short subject and object who-questions
in German-speaking CHI with mild to severe HL aged 3 to 4 years (Wimmer, Rothweiler, and
Penke, 2017; Penke and Wimmer, 2018). They employed a questions-after-story task in which
the childrenwere shownapicture of twopeople interacting, thenheard a short story consisting
of three sentences, and finally had to answer a question about the picture by pointing to the
corresponding person in it. Both studies report that the CHI were overall significantly less
accurate than age-matched CTD.When looking at group differences in each question type, the
results of these studies differed: Wimmer et al. (2017) found no significant interaction between
hearing status and question type, but Penke and Wimmer (2018) found that the two groups
differed significantly on object questions and marginally on subject questions. Regarding the
comparison between question types, the two studies report that both groups comprehended
subject who-questions better than object who-questions; however, this asymmetry was more
pronounced in the CHI group (Wimmer et al., 2017) and not significant in the CTD group (Penke
& Wimmer, 2018).
In relation to the stages of acquisition of these structures, Wimmer et al. (2017) found that,
with object questions, CHI aged 3 performed at chance level, while CHI aged 4 performed sim-
ilarly to 3-year-old CTD. Moreover, the authors observed three different performance patterns
present – with different proportions – in both groups of children: chance performance with
both subject and object who-questions, above chance performance with subject questions,
and above chance performance with both types of who-questions. The latter, more advanced
pattern was reached at age 3 by most CTD (73% at age 3, 92% at age 4) but only at age 4 by
most CHI (22% at age 3, 83% at age 4). Considering also that there was no significant inter-
action between hearing status and age, the authors suggested that the CHI showed the same
stages of acquisition as the CTD, but delayed. Penke and Wimmer (2018) tested 10 children in
their CHI group again when they were 6-8 years old. For this subgroup of CHI, mean accuracy
13
at first testing was 92% and 82.8% for subject and object who-questions, respectively, and did
not differ significantly from the rest of the group. At retest, their mean accuracy with both
types ofwho-questions had improved to 99% and all 10 children had overall accuracy levels of
95% or higher.
Ruigendijk and Friedmann (2017) examined the comprehension and repetition of subject
and object who- and which-questions in German-speaking CHI with moderate to profound
HL aged 9-13 years. In a point-in-picture task, CHI showed worse comprehension than CTD
aged as the younger CHI for subjectwhich-questions (significant difference) and objectwhich-
questions (marginal difference, which the authors attributed to some CTD still having diffi-
culties with this structure). While the CTD group performed above chance on all four types
of questions, the CHI group did so on subject who- and which-questions and on object who-
questions, but they performed at chance level on objectwhich-questions. In contrast to Penke
and Wimmer (2018), in this study the subject/object asymmetry in comprehension was signif-
icant in both groups of children and with both types of wh-questions.
Ruigendijk and Friedmann (2017) also report on CHI’s question repetition with two tasks:
the first task included long subject and object who-questions (with an extra prepositional
phrase) and short subject and object which-questions; the second one included long subject
and object who- and which-questions. In the first repetition task, the CHI performed similarly
to aged-matched CTD on both types of subject questions (i.e. long subjectwho-questions and
short subjectwhich-questions), and significantly below the CTD group on both types of object
questions (long object who-questions and short object which-questions). In the second repe-
tition task, the CTD group was not age-matched to the CHI but was aged as the younger CHI;
there were no significant differences between groups on any of the four types of long ques-
tions. The authors indicate that this result might be due to the CTD being younger and the
sentences longer.
Another work on the comprehension ofwh-questions by German-speaking CHI was carried
out by Schouwenaars et al. (2019), who investigated not only how accurate these children’s
comprehension was but also how they processed these structures. A picture selection task
with eye-tracking and assessing subject, object and passive which-questions was employed in
a group of CHI who had severe to profound HL and were 7 to 12 years old. The study also
examined whether performance varied when different cues were available for interpretation.
14
In passive which-questions, the interpretation of the first noun phrase (NP) was always dis-
ambiguated by the verb in passive voice and the by-phrase. For subject and object which-
questions, there were three different disambiguation conditions: disambiguation by the case
of the first NP which contained the wh-word, by verb agreement, or by both case and agree-
ment. The results of the questions task reported on included only those children who showed
good perception of case and agreement information in two screening tasks. Both in accuracy
and gaze patterns, the CHI performed similarly to a group of CTD matched on hearing age (i.e.
aged 7 to 10) on subject and passive questions but differently on object questions. With sub-
ject and passive questions both CHI and CTD performed at ceiling and looked initially more at
the competitor and, then, more at the target. With object questions, both groups of children
were less accurate, but CHI comprehended them significantly worse than CTD. Furthermore,
CHI’s looks towards the target increased later and less steeply than for CTD, which was inter-
preted by the authors as indicating that the CHI were slower in revising their initial agent-first
interpretation of object questions. The type of disambiguation did not seem to have an effect
on accuracy, but it did on gaze patterns: while the CTD showed an initial subject interpreta-
tion on object questions (seen in an initial preference for the competitor, agent-first picture)
when theywere disambiguated by agreement alone or by agreement and case, but not by case
alone, CHI also showed an initial subject interpretation on object questions disambiguated by
case, which are disambiguated earlier. However, the fact that looks towards the agent-first
picture differed between subject and object questions before the onset of the verb indicates
that the CHI did not completely ignore the early case cue.
DeLuca (2015) also investigated the gaze patterns during who- and which-question com-
prehension, in English-speaking CHI with CIs. In this eye-tracking study, CHI aged 7-11 years
listened to a context sentence and a question, and had to answer by clicking on one of four
pictures, each showing a different character. The questions included subject and object who-
and which-questions with low and high memory load; questions with high memory load were
made longer by adding adjectives in the NPs. Overall, the CHI were significantly less accurate in
their responses than a group of age-matched CTD. In trials answered correctly, the CHI did not
differ significantly from the CTD in their fixations to the target picture during the syntactic gap.
However, in the CHI group the probability of fixating the target was smaller with object than
with subject questions - although reported as a tendency. In trials answered incorrectly, gaze
patterns during the gap did differ significantly between the two groups. For subject questions,
15
both groups fixated more at the object, which suggests they were both reversing the thematic
roles. When they were not looking at the object, the CTD fixated at the other three pictures
in equal proportion, but the CHI looked more at the description foil (i.e., an unrelated char-
acter that shared one descriptive similarity with the target, for instance both were female or
both were elderly people), followed by the target and then by the attribute-related foil to the
target noun phrase (i.e. the same picture with the defining attribute changed). On incorrectly
answered object questions, the CTD exhibited the highest proportion of fixations at the target,
followed by the three foils with very similar proportions. In contrast, the CHImost often fixated
the attribute-related object foil, followed by the subject, then the target picture and, finally,
the description foil. It seems that, for incorrectly answered object questions, the CTD might
have originally assigned thematic roles correctly even if they later failed to select the target
picture; in the case of CHI, for whom the highest competition was from the same noun phrase
category as the target, incorrect responses might have resulted from working memory limita-
tions. Indeed, the CHI performed more poorly than the CTD on a battery of working memory
tasks that was administered as part of the same study.
Volpato and D’Ortenzio (2018) and D’Ortenzio and Volpato (2020) studied the production
of subject and object who- and which-questions in Italian-speaking CHI with CIs. Both studies
employed the same type of elicitation task used by Friedmann and Szterman (2011) and in-
cluded CHI of very similar ages: 7-13 and 7-12, respectively. Volpato and D’Ortenzio (2018) did
not find significant differences between the CHI and a group of age-matched CTD in accuracy
nor in the use of the different correct and incorrect answering strategies. In contrast, the CHI
in D’Ortenzio and Volpato (2020) performed significantly below a group of age-matched CTD
and another group of CTD matched on hearing age since HA fitting2 on object who-questions
and subject and object which-questions; on subject who-questions, however, they performed
at ceiling. The differences between the CHI and the two CTD groupswere significant when con-
sidering all types of correct responses and only wh-V-N target structures (i.e., who-verb-noun
phrase for who-questions or which+noun phrase-verb-noun phrase for which-questions), but
not when considering all the other correct structures together (e.g., with topicalized or omit-
ted subject, passive questions, etc.). The CHI also produced significantly more ungrammatical2CHI are usually first fitted with HAs and, if these do not provide enough improvement, then implanted with
CIs. Nonetheless, hearing age for CHI with CIs is generally calculated as the time since the first CI activation, not
since HA fitting, as was done in this case.
16
or incomplete sentences than both CTD groups.
The different results found by these two studiesmight be related to the clinical characteris-
tics of the CHI groups. All the CHI in D’Ortenzio and Volpato (2020) were diagnosed and started
intervention before their first year of life, whereas for the CHI in Volpato and D’Ortenzio (2018)
this happened later, between birth and the age of 3;6 years. In the latter study the authors do
not provide information on the CHI’s degree of hearing loss and, while we can assume it was
severe or profound because they all had at least one CI, it is possible that these children were
diagnosed later because their hearing loss was less severe and they hadmore residual hearing.
If this was the case, it might explain why they performed similarly to controls and the CHI in
D’Ortenzio and Volpato (2020) did not.
There is, in all, convergent evidence that wh-questions pose considerable difficulties for
CHI, both in comprehension and production. Differences between CHI and CTD have been ob-
served not only in offline measures (accuracy) but also in online measures (gaze patterns). In
general, the gradient of difficulty that emerges is that who-questions are easier than which-
questions and, within each type, object questions are harder, with objectwhich-questions be-
ing the hardest of all. Furthermore, CHI are reported to produce more ungrammatical sen-
tences than CTD. However, not all evidence points to trouble with the same types of wh-
questions. While some studies report no differences between CHI and age-matched CTD, oth-
ers have found CHI to perform worse than even much younger groups of CTD. Further investi-
gation is needed to understand why different studies obtained different results. In particular,
future studies might gain insight from looking into the relation betweenwh-question compre-
hension and production and cognitive abilities such as working memory.
Sources of variability
If language acquisition in CTD is characterized by wide variability (Kidd & Donnelly, 2020),
language acquisition in hearing-impaired children is even more so. A long list of additional
factors come into play: degree of HL and level of residual hearing; age at onset of HL; cause
of HL; type of hearing device; duration of hearing device use; type, duration, and frequency
of speech therapy; parental hearing status; mode of communication used at home and at the
educational setting (see Tomblin et al., 2015 for CHI with mild to severe HL with HAs and Gillis,
2018 for a review focusing on CHI with CIs). Authors usually account for these differences by
17
providing precise descriptions of the CHI group and, often, also of each child. Additionally,
many authors carry out individual analyses comparing the performance of each child in the
CHI group to that of the control group and/or to chance level.
These analyses have revealed that, despite significant differences between CHI and CTD at
the group level, some CHI perform well even in the most problematic structures. For example,
in Delage (2008), 2 of the 14 French-speaking CHI aged 7-11 performed at ceiling on the pro-
duction of object relatives in the elicitation task. In Ruigendijk and Friedmann (2017), 12 out
of 19 German-speaking CHI showed comprehension similar to the age-matched control group
on object relatives, 10 and 12 out of 16 on subject and object which-questions, respectively,
and 10 out of 16 on topicalized sentences. Of the 15 CHI who participated in the wh-question
repetition task, those who performed similarly to age-matched CTD were 12 and 8 for long
subject and object who-questions, respectively, and 14 and 10 for short subject and object
which-questions, respectively. In Schouwenaars et al. (2019), 13 out of 21 German-speaking
CHI interpreted object which-questions with accuracy levels of 70% and higher, and two of
them reached 100%.
As can be seen from these examples, group results often mask individual variability, which
is why individual analyses are a good and necessary practice when studying this population.
In the section below, we briefly comment on different extralinguistic factors that could
modulate individual performance of this group of children.
Demographic, clinical and cognitive factors related to CHI’s sentence comprehension and
production
Some of the studies that looked into CHI’s sentence comprehension and production by target-
ing specific structures (instead of overall performance on a range of diverse structures) have
also explored the influence of various demographic and clinical factors on their performance.
Among the demographic factors, age was found to correlate positively with the comprehen-
sion of sentences with subject relative clauses (Volpato, 2010) and of object who-questions
(Penke & Wimmer, 2018), and to be a significant predictor of which-question comprehension
(Schouwenaars et al., 2019), but no correlation was found between age and overall relative
clause comprehension (Davis and Blasdell, 1975; Friedmann and Haddad-Hanna, 2014), com-
prehension of object relative clauses with preverbal or postverbal subject (Volpato, 2010),
18
production of relative clauses and rate of complex utterances (i.e. with one or more embed-
ded clauses) in spontaneous speech (Delage, 2008), comprehension of which-questions and
topicalization structures (Friedmann & Haddad-Hanna, 2014), comprehension of subjectwho-
questions (Penke &Wimmer, 2018), and eye fixations to the target picture at the syntactic gap
during who- and which-question comprehension (DeLuca, 2015). Hearing age was also found
to be a significant predictor of which-question comprehension, but it did not correlate with
subject, object and overall who-question comprehension (Penke & Wimmer, 2018).
Evidence inquiring the relation between sentence comprehension and production and clin-
ical hearing loss-related variables (e.g. degree of aided and unaided hearing loss, age of HA
fitting, age of implantation, type of hearing device, age of diagnosis, age of onset of inter-
vention, length of speech therapy) shows mixed results (see Tables B1 and B2 in Appendix B).
While most studies found no significant results, a few studies have shown significant correla-
tion among these factors. For instance, Delage (2008) reported a negative correlation between
the rate of complex utterances in spontaneous speech and the degree of unaided hearing loss.
Children with more severe HL produced significantly fewer complex utterances. A positive cor-
relation between the spontaneous production of relative clauses without a main clause (e.g.
The girl who plays) and the length of speech therapy was also found. Longer CI use was related
to better production of subject relatives in Volpato and Vernice (2014) and longer use of HAs
prior to cochlear implantation was related to better production of object which-questions in
D’Ortenzio and Volpato (2020). Age at HA fitting correlated negatively with comprehension of
object relatives and topicalization structures with OVS order (pooled together) in Friedmann
and Szterman (2006), i.e. the younger the child was at HA fitting, the higher her chances for
having average comprehension levels above 70%. Additionally, having had HAs fitted before 8
months was significantly correlated with having more chances of good comprehension (aver-
age accuracy above 70%). Age at HA fitting also correlated negatively with production of object
who-questions in D’Ortenzio and Volpato (2020). Since the clinical factors mentioned here are
related, e.g. earlier fitting with hearing devices results in longer hearing device use, further
research is needed to shed light on which of them are actually influencing CHI’s performance.
Apart from demographic and clinical factors, some studies have looked into cognitive fac-
tors (see Table B3 in Appendix B). Few of the studies reviewed here have analyzed the relation
between sentence comprehension and production and nonverbal abilities: a positive corre-
lation with relative clause production in spontaneous speech was reported in Delage (2008),
19
but no significant correlations were found with who-question comprehension in Penke and
Wimmer (2018). A cognitive factor that has recently received increasing attention is memory
capacity, in particular short-termmemory andworkingmemory. Lee et al. (2018) reported that
passive sentence comprehension was highly correlated to working memory capacity as mea-
sured by a composite score including forward and backward digit and word spans (all admin-
istered in their visual versions), and that working memory accounted for 27.1% of the variance
in passive sentence comprehension scores. Higher forward digit span was related to better
comprehension of object relatives with preverbal subject with number mismatch between the
first and the second NPs and of object relatives with postverbal subject in Volpato (2020). In
Volpato (2010), forward digit span was not significantly correlated with relative clause com-
prehension, but forward word span was. Neither Volpato (2020) nor Volpato (2010) found
significant correlations with backward digit span. In contrast, Schouwenaars et al. (2019) did
report that better objectwhich-question comprehension was related to higher backward digit
span, although the lack of an interaction with group indicates that this was the case for CTD
too. Again, it is hard to draw conclusions from these results. Although there is an increas-
ing amount of evidence of the importance of memory resources for CHI’s language outcomes
(Pisoni and Geers, 2000; Dillon, Cleary, Pisoni, and Carter, 2004; Harris et al., 2013), it is still
not clear whether memory capacity is specifically associated to sentence comprehension and
production.
Accounts of CHI’s difficulties in sentence comprehension and production
Different explanations of CHI’s morphosyntactic difficulties have been proposed, accounting
for poor linguistic attainment either as a direct result of hearing loss or as an indirect result
of other aspects of cognitive development. Here we briefly review four accounts that point,
respectively, to perceptual limitations, insufficient exposure to language early in life, and im-
paired memory or impaired cognitive sequencing caused by insufficient exposure to sound
early in life.
There is abundant evidence from various languages that CHI often omit or are less ac-
curate than CTD in their use of verbal inflection, nominal inflection, agreement, determin-
ers, pronouns, prepositions, and other grammatical morphemes (e.g. Szagun, 2000, 2004;
Le Normand, Ouellet, and Cohen, 2003; Hammer, 2010; Guo, Spencer, and Tomblin, 2013;
20
Koehlinger, Horne, and Moeller, 2013; Guasti et al., 2014; Sundström, Löfkvist, Lyxell, and
Samuelsson, 2018; Werfel, 2018). Some authors (Szagun, 2000, 2004; Svirsky, Stallings, Lento,
Ying, and Leonard, 2002; Caselli et al., 2012) have suggested that these difficulties are related
to CHI’s perceptual limitations. Since grammatical markers are often short, formed by high fre-
quency sounds, and unstressed, they are perceptually less salient and thus harder to perceive.
Trouble perceiving grammatical morphemes would have consequences for the acquisition of
grammar and the comprehension and production of sentences. Support for the perceptual
salience hypothesis has been provided, for instance, by Svirsky et al. (2002), who showed that
the order of acquisition of noun plural, copula, and past tense in English-speaking CHI was dif-
ferent than in CTD and that perceptually more salient elements were acquired earlier. Szagun
(2000) combined this perceptual salience hypothesis with the idea that CHI probably have dif-
ficulties with the temporal processing of auditory stimuli. She proposed that, because CHI have
difficulties discriminating speech sounds, they might fail to process speech sufficiently and, as
a result, focus on content words instead, missing crucial grammatical information. In contrast
to simple sentences where the overall meaning can be conveyed by content words, missing
grammatical cues (be it because of low perceptual salience or because of insufficient audi-
tory processing) would be particularly serious in the case of sentences that involve complex
structures, like those reviewed in this work.
While the perceptual salience account focuses on the auditory limitations that CHI may
experience even after they have received proper hearing devices and language therapy, other
accounts focus on the period during which CHI did not receive (sufficient) auditory input. On
the one hand, several authors (Friedmann and Szterman, 2006; Delage and Tuller, 2007; Fried-
mann and Szterman, 2011; Friedmann and Haddad-Hanna, 2014; Ruigendijk and Friedmann,
2017) attribute CHI’s morphosyntactic difficulties to the lack of sufficient exposure to natu-
ral language during a critical (Lenneberg, 1967) or sensitive period for language acquisition.
Namely, this account posits that being exposed to restricted and impoverished input during
early childhood as a result of hearing loss hinders the development of morphosyntactic ability
in CHI. In this account, exposure to language during the critical/sensitive period is viewed as
a necessary, but not sufficient, condition for language development. However, it is still un-
clear the length of the critical/sensitive period and the exact type of linguistic input that is
needed during said period for typical morphosyntactic development in general, and for the
acquisition of complex sentences in particular (a similar concern is raised in Ruigendijk and
21
Friedmann, 2017). Another open question is how the lack of adequate linguistic input leads
to morphosyntactic deficit, i.e. which are the linguistic mechanisms or abilities that are al-
tered, and how. In the case of CHI who are able to perceive morphosyntactic cues but fail to
use them in comprehension of complex sentences, Schouwenaars et al. (2019) have proposed
that insufficient quality and quantity of linguistic input during development may lead to less
automatized, slower processing, which may contribute to CHI’s morphosyntactic difficulties.
On the other hand, it has been proposed that it is not the lack of exposure to language, but
the lack of exposure to auditory input in early childhood which indirectly causes CHI’s mor-
phosyntactic difficulties. Two different accounts of this indirect link have been put forward.
The first one posits that CHI’s morphosyntactic difficulties arise because of immature or atyp-
ically developed performance systems, which are a consequence of hearing loss during early
development (Delage, 2008; Tuller and Delage, 2014). Due to degraded auditory input, infants
with hearing loss would need to allocate more attentional resources to sound processing than
CTD, leaving fewer resources for other cognitive processes such as short-term memory and
working memory; as a consequence, these performance systems would not mature normally.
Since linguistic structures of high computational complexity like the sentences mentioned in
thiswork put high demands onworkingmemory, CHIwould encounter difficulties in processing
and acquiring them, leading to their morphosyntactic difficulties.
The other account of the indirect link between hearing loss and morphosyntactic difficul-
ties is the auditory scaffolding hypothesis (Conway, Pisoni, & Kronenberger, 2009), which holds
that exposure to sound, given its temporal and sequential nature, provides a scaffolding for
learning how to process sequential information. Auditory deprivation early in life would then
affect the development of domain-general cognitive sequencing capacities, which are neces-
sary for acquiring spoken language, because they allow to extract structural regularities from
the linguistic input (Conway et al., 2009, p. 277). Such an account raises the question of
whether CHI’s language skills can be improved by enhancing general cognitive abilities such
as sequencing and statistical learning. Bedoin et al. (2018) observed that CHI with CIs showed
improvement in a grammaticality judgement task and in a non-word repetition task after ses-
sions of morphosyntactic training with musical primes, but not after training sessions where
the primes were environmental sounds without rhythmic structure. However, this difference
was not observed in a syntax comprehension task in which the participants were required to
follow instructions to execute actions; in this case, performance improved after training with
22
both rhythmic and non-rhythmic primes. Although there is evidence supporting the auditory
scaffolding hypothesis, there is also evidence inconsistent with it (see Deocampo, Smith, Kro-
nenberger, Pisoni, and Conway, 2018 for a review). Furthermore, the direction of the relation-
ship between language and cognitive sequencing in development is also under debate, and it
has been proposed that poor sequencing abilities might not cause but be caused by poor lan-
guage development (language scaffolding hypothesis, Hall, Eigsti, Bortfeld, and Lillo-Martin,
2018). Further research is needed to shed light on the link between language and sequencing
abilities in development, as well as on the potential of cognitive training for improving spoken
language skills in CHI.
Given the wide variability present in CHI, neither of these accounts can explain all cases of
morphosyntactic difficulties in CHI on its own. For example, not all CHI who showmorphosyn-
tactic difficulties fail to perceive less salient grammatical markers (Ruigendijk and Friedmann,
2017; Schouwenaars et al., 2019), and some studies have failed to find correlations between
measures of short-termmemory and working memory capacity, and performance in sentence
comprehension and production (Volpato, 2010; Penke and Wimmer, 2018). Most likely, all of
the phenomena described by these accounts are relevant factors, and just how relevant each
of them is differs from one case to another, depending on the interplay with other factors.
Besides how the morphosyntactic difficulties in CHI can be explained, it is also a matter of
debate whether these difficulties are indicative of a persistent deficit or of a delay in linguistic
development. Proponents of the persistent deficit hypothesis argue that evidence of CHI’s dif-
ficulties in adolescence suggests that language does not generally normalize with age (Delage
and Tuller, 2007; Friedmann and Haddad-Hanna, 2014). However, longitudinal studies on sen-
tence comprehension and production that would allow to clarify this issue are extremely rare.
Delage (2008), also reported in Tuller and Delage (2014), analyzed the spontaneous speech
of a group of CHI with mild to moderate HL at ages 6;1-11;11 and then again at ages 7;11-13;11.
They found that only the youngest children (mean age of 8 at first testing and of 10 at sec-
ond testing) showed improvements in mean length of utterance and rate of embedding, while
the eldest children (mean age of 11 at first testing and of 13 at second testing) stagnated and
still performed below a group of 11-year-old TDC. Opposite to this finding, Penke and Wimmer
(2018) assessed who-question comprehension in a group of CHI with mild to severe HL aged
3-4 years, and then again in a subset of those CHI when they were 6-8 years. While at age 3-4
the CHI performed significantly below age-matched CTD, at age 6-8 they performed at ceiling.
23
As has been pointed out by Penke and Wimmer (2018, p. 14), more longitudinal studies test-
ing a variety of syntactic structures and, we might add, a wider age range, are needed to help
settle the debate whether CHI’s difficulties evince a persistent deficit or a delay.
Conclusions
In this work we have reviewed the evidence on how children with prelingual hearing loss who
use spoken language comprehend and produce sentences, focusing on simple sentences, top-
icalized sentences, passives, sentences with relative clauses, and wh-questions. We have also
gone over factors that introduce heterogeneity in the population of CHI and revised the ev-
idence on how their performance with specific sentence structures relates to demographic,
clinical and cognitive variables. Finally, we have summarised different accounts that have tried
to explain the morphosyntactic difficulties that characterise this population.
Studies have shown overall good performancewith simple, active sentenceswith canonical
word order. In contrast, CHI display visible difficulties with more complex sentences, specially
with topicalization, relative clauses and wh-questions. More studies are necessary to deter-
minewhether passive sentences are also generally problematic. We encourage future research
on CHI’s language development to expand the range of languages and structures tested, while
considering the specificities of each language and analyzing performance in each type of struc-
ture separately. This will help pinpoint the difficulties experienced by CHI, which will in turn
improve therapeutic and educational practices.
Better comprehension and production of complex sentences has been related to earlier
intervention and longer use of hearing devices, as well as to higher short-term memory and
working memory capacity. While the relation with many other variables has been assessed,
the results are not convergent. What is clear, however, is that the heterogeneity characteristic
of the CHI population must be considered carefully, and that sources of heterogeneity need to
be attended. We recommend carrying out individual analyses along with group analyses.
This literature review reveals two areas of research that need further development. First,
studies of online processing of sentences in CHI are very scarce. We believe these are neces-
sary to obtain a deeper understanding of how CHI acquire and process spoken language, the
time-course of morphosyntactic cues retrieval and its relation to other domain-general capac-
24
ities like working memory capacity, planning and self-monitoring. Second, longitudinal studies
assessing how sentence comprehension and production evolve as CHI grow up are also very
rare. This might be partly due to the fact thatmodern hearing devices are relatively recent. We
expect that, as early hearing loss detection programs become more frequent worldwide and
more children with HL access modern hearing devices and therapy, it will be easier to conduct
longitudinal studies. This kind of studies will hopefully shed light on the development of CHI’s
morphosyntactic difficulties.
References
Bedoin, N., Besombes, A.-M., Escande, E., Dumont, A., Lalitte, P., & Tillmann, B. (2018). Boost-
ing syntax training with temporally regular musical primes in children with cochlear im-
plants. Annals of Physical and Rehabilitation Medicine, 61(6), 365–371. doi:10.1016/j.
rehab.2017.03.004
Bishop, D. V. M. (2003). Test for reception of grammar (trog-2) (2nd ed.). San Antonio, TX: The
Psychological Corporation.
Boons, T., de Raeve, L., Langereis, M., Peeraer, L., Wouters, J., & van Wieringen, A. (2013). Ex-
pressive vocabulary, morphology, syntax and narrative skills in profoundly deaf children
after early cochlear implantation. Research in Developmental Disabilities, 34(6), 2008–
2022. doi:10.1016/j.ridd.2013.03.003
Briscoe, J., Bishop, D. V. M., & Norbury, C. F. (2001). Phonological processing, language, and
literacy: A comparison of children with mild-to-moderate sensorineural hearing loss and
those with specific language impairment. Journal of Child Psychology and Psychiatry and
Allied Disciplines, 42(3), 329–340. doi:10.1017/S0021963001007041
Caselli, M. C., Rinaldi, P., Varuzza, C., Giuliani, A., & Burdod, S. (2012). Cochlear implant in the
second year of life: Lexical and grammatical outcomes. Journal of Speech, Language, and
Hearing Research, 55, 382–394. doi:10.1044/1092-4388(2011/10-0248)
Conway, C. M., Pisoni, D. B., & Kronenberger, W. G. (2009). The importance of sound for cogni-
tive sequencing abilities. the auditory scaffolding hypothesis. Current Directions in Psy-
chological Science, 18(5), 275–279. doi:10.1111/j.1467-8721.2009.01651.x
25
Costa, J. & Moita, M. (2016). Movimento x0 em crianças com implante coclear: Evidência para
défice sintático em casos de input tardio. Revista da Associação Portuguesa de Linguís-
tica, 2, 205–234. doi:10.21747/2183-9077/rapl2a9
Dammeyer, J. (2010). Psychosocial development in aDanish population of childrenwith cochlear
implants and deaf and hard-of-hearing children. Journal of Deaf Studies and Deaf Edu-
cation, 15(1), 50–58. doi:10.1093/deafed/enp024
Davis, J. & Blasdell, R. (1975). Perceptual strategies employed by normal-hearing and hearing-
impaired children in the comprehension of sentences containing relative clauses. Journal
of Speech and Hearing Research, 18(2), 281–295. doi:10.1044/jshr.1802.281
de Villiers, P. A. (1988). Assessing English syntax in hearing-impaired children: Elicited produc-
tion in pragmatically motivated situations. Journal of the Academy of Rehabilitative Au-
diology, 21(Mono Suppl), 41–71.
Delage, H. (2008). Evolution de l’hétérogénéité linguistique chez les enfants sourds légers et
moyens: Etude de la complexité syntaxique (Doctoral dissertation, François-Rabelais Uni-
versity, Tours).
Delage, H. & Tuller, L. (2007). Language development andmild-to-moderate hearing loss: Does
language normalize with age? Journal of Speech, Language, and Hearing Research, 50,
1300–1313. doi:10.1044/1092-4388(2007/091)
DeLuca, Z. W. (2015).Wh-question processing in children with cochlear implants (Doctoral dis-
sertation, CityUniversity ofNewYork, NewYork). Retrieved fromhttps://academicworks.
cuny.edu/gc_etds/902
Deocampo, J. A., Smith, G. N. L., Kronenberger, W. G., Pisoni, D. B., & Conway, C. M. (2018). The
role of statistical learning in understanding and treating spoken language outcomes in
deaf childrenwith cochlear implants. Language, Speech, and Hearing Services in Schools,
49, 723–739. doi:10.1044/2018_LSHSS-STLT1-17-0138
Dillon, C. M., Cleary, M., Pisoni, D. B., & Carter, A. K. (2004). Imitation of nonwords by hearing-
impaired children with cochlear implants: Segmental analyses. Clinical Linguistics & Pho-
netics, 18(1), 39–55. doi:10.1080/0269920031000151669
D’Ortenzio, S. & Volpato, F. (2020). How do Italian-speaking children handle wh-questions? a
comparison between children with hearing loss and children with normal hearing. Clini-
cal Linguistics & Phonetics, 34(4), 407–429. doi:10.1080/02699206.2019.1677779
26
Duchesne, L., Sutton, A., & Bergeron, F. (2009). Language achievement in children who re-
ceived cochlear implants between 1 and 2 years of age: Group trends and individual pat-
terns. Journal of Deaf Studies and Deaf Education, 14(4), 465–485. doi:10.1093/deafed/
enp010
Friedmann, N. & Costa, J. (2011). Last resort and no resort: Resumptive pronouns in Hebrew
and Palestinian Arabic hearing impairment. In A. Rouveret (Ed.), Resumptive pronouns
at the interfaces (pp. 223–239). John Benjamins. doi:10.1075/lfab.5
Friedmann, N. & Haddad-Hanna, M. (2014). The comprehension of sentences derived by syn-
tactic movement in Palestinian Arabic-speaking children with hearing impairment. Ap-
plied Psycholinguistics, 35(3), 473–513. doi:10.1017/S0142716412000483
Friedmann, N., Novogrodsky, R., Szterman, R., & Preminger, O. (2008). Resumptive pronouns
as last resort when movement is impaired: Relative clauses in hearing impairment. In
S. Armon-Lotem, S. Rothstein, & G. Danon (Eds.), Current issues in generative hebrew
linguistics (pp. 276–290). Amsterdam, The Netherlands: John Benjamins. doi:10.1075/
la.134.11res
Friedmann, N., Rizzi, L., & Belletti, A. (2016). No case for case in locality: Case does not help
interpretation when intervention blocks a-bar chains. Glossa: a journal of general lin-
guistics, 2(1), 33. doi:10.5334/gjgl.165
Friedmann, N. & Szterman, R. (2006). Syntactic movement in orally-trained childrenwith hear-
ing impairment. Journal of Deaf Studies and Deaf Education, 11, 56–75. doi:10 . 1093/
deafed/enj002
Friedmann, N. & Szterman, R. (2011). The comprehension and production of wh-questions in
deaf and hard-of-hearing children. Journal of Deaf Studies and Deaf Education, 16(2),
212–235. doi:10.1093/deafed/ENQ052
Fujiyoshi, A., Fukushima, K., Taguchi, T., Omori, K., Kasai, N., Nishio, S., . . . Shiroma, M. (2012).
Syntactic development in Japanese hearing-impaired children. Annals of Otology, Rhi-
nology & Laryngology, 121(4), 28–34. doi:10.1177/000348941212100404
Geers, A. E., Moog, J. S., Biedenstein, J., Brenner, C., & Hayes, H. (2009). Spoken language
scores of children using cochlear implants compared to hearing age-mates at school en-
try. Journal of Deaf Studies and Deaf Education, 14(3), 371–385. doi:10.1093/deafed/
enn046
27
Geers, A. E., Nicholas, J. G., & Sedey, A. L. (2003). Language skills of childrenwith early cochlear
implantation. Ear and Hearing, 24, 46S–58S. doi:10.1097/01.AUD.0000051689.57380.
1B
Gillis, S. (2018). Speech and language in congenitally deaf childrenwith a cochlear implant. In A.
Bar-On & D. Ravid (Eds.),Handbook of communications disorders: Theoretical, empirical,
and applied linguistic perspectives (pp. 765–792). Berlin: De Gruyter Mouton. doi:10 .
1515/9781614514909-038
Guasti, M. T., Papagno, C., Vernice, M., Cecchetto, C., Giuliani, A., & Burdo, S. (2014). The effect
of language structure on linguistic strengths and weaknesses in children with cochlear
implants: Evidence from Italian. Applied Psycholinguistics, 35, 739–764. doi:10 .1017/
S0142716412000562
Guo, L.-Y., Spencer, L. J., & Tomblin, B. (2013). Acquisition of tense marking in English-speaking
children with cochlear implants: A longitudinal study. Journal of Deaf Studies and Deaf
Education, 18(2), 187–205. doi:10.1093/deafed/ens069
Hall, M. L., Eigsti, I.-M., Bortfeld, H., & Lillo-Martin, D. (2018). Auditory access, language access,
and implicit sequence learning in deaf children. Developmental Science, 21(3), e12575.
doi:10.1111/desc.12575
Halliday, L. F. & Bishop, D. V. M. (2005). Frequency discrimination and literacy skills in chil-
dren with mild to moderate sensorineural hearing loss. Journal of Speech, Language,
and Hearing Research, 48, 1187–1203. doi:10.1044/1092-4388(2005/083)
Halliday, L. F., Tuomainen, O., & Rosen, S. (2017). Language development and impairment in
children with mild to moderate sensorineural hearing loss. Journal of Speech, Language,
and Hearing Research, 60, 1551–1567. doi:10.1044/2016_JSLHR-L-16-0297
Hammer, A. (2010). The acquisition of verbal morphology in coclear-implanted and specific
language impaired children (Doctoral dissertation, Faculteit der Letteren, Leiden Uni-
versity). Retrieved from https://www.lotpublications.nl/the-acquisition-of-verbal-
morphology-in-cochlear-implanted-and-specifi-the-acquisition-of-verbal-morphology-
in-cochlear-implanted-and-specific-language-impaired-children
Hansson, K., Forsberg, J., Löfqvist, A., Mäki-Torkko, E., & Sahlén, B. (2004). Working memory
and novel word learning in children with hearing impairment and children with specific
language impairment. International Journal of Language & Communication Disorders,
39(3), 401–422. doi:10.1080/13682820410001669887
28
Harris, M. S., Kronenberger, W. G., Gao, S., Hoen, H. M., Miyamoto, R. T., & Pisoni, D. B. (2013).
Verbal short-termmemory development and spoken language outcomes in deaf children
with cochlear implants. Ear andHearing, 34(2), 179–192. doi:10.1097/AUD.0b013e318269ce50
Inscoe, J. R., Odell, A., Archbold, S., & Nikolopoulos, T. (2009). Expressive spoken language de-
velopment in deaf children with cochlear implants who are beginning formal education.
Deafness & Education International, 11(1), 39–55. doi:10.1002/dei.252
Kidd, E. & Donnelly, S. (2020). Individual differences in first language acquisition. Annual Re-
view of Linguistics, 6(1), 319–340. doi:10.1146/annurev-linguistics-011619-030326
Kitzinger,M.& John, J. (1980). Hearing-impaired children’s understanding of active, passive and
scrambled sentences. In I. G. Taylor & A. Markides (Eds.), Disorders of auditory function
(Vol. 3, pp. 73–78). Academic Press.
Koehlinger, K. M., Horne, A. J. O. V., & Moeller, M. P. (2013). Grammatical outcomes of 3- and
6-year-old children who are hard of hearing. Journal of Speech, Language, and Hearing
Research, 56(5), 1701–1714. doi:10.1044/1092-4388(2013/12-0188)
Le Normand,M.-T., Ouellet, C., & Cohen, H. (2003). Productivity of lexical categories in French-
speaking children with cochlear implants. Brain and Cognition, 53(2), 257–262. doi:10.
1016/S0278-2626(03)00122-2
Lee, Y., Sung, J. E., & Sim, H. (2018). Passive sentence comprehension difficulties and its related
factors in children with cochlear implants. International Journal of Pediatric Otorhino-
laryngology, 109, 60–66. doi:10.1016/j.ijporl.2018.03.02
Lenneberg, E. H. (1967). Biological foundations of language. New York: Wiley.
Mandal, J. C., Kumar, S., & Roy, S. (2016). Comparison of auditory comprehension skills in chil-
dren with cochlear implant and typically developing children. International Journal of
Pediatric Otorhinolaryngology, 91, 113–120. doi:10.1016/j.ijporl.2016.10.009
Moita, M. & Lobo, M. (2018). Compreensão e produção oral de interrogativas-q em crianças
portuguesas surdas com implante coclear. Revista da Associação Portuguesa de Linguís-
tica, 4, 168–189. doi:10.26334/2183-9077/rapln4ano2018a39
Nikolopoulos, T. P., Dyar, D., Archbold, S., & O’Donoghue, G. M. (2004). Development of spo-
ken language grammar following cochlear implantation in prelingually deaf children.
Archives of Otolaryngology - Head and Neck Surgery, 130(5), 629–633. doi:10 . 1001 /
archotol.130.5.629
29
Niparko, J. K., Tobey, E. A., Thal, D. J., Eisenberg, L. S., Wang, N.-Y., Quittner, A. L., & Fink, N. E.
(2010). Spoken language development in children following cochlear implantation. Jour-
nal of the American Medical Association, 303(15), 1498–1506. doi:10.1001/jama.2010.
451
Nittrouer, S., Sansom, E., Low, K., Rice, C., & Caldwell-Tarr, A. (2014). Language structures used
by kindergartners with cochlear implants: Relationship to phonological awareness, lex-
ical knowledge and hearing loss. Ear and Hearing, 35(5), 506–518. doi:10.1097/AUD.
0000000000000051
Norbury, C. F., Bishop, D. V. M., & Briscoe, J. (2001). Production of English finite verb mor-
phology: A comparison of sli andmild–moderate hearing impairment. Journal of Speech,
Language, and Hearing Research, 44, 165–178. doi:1092-4388/01/4401-0165
Norbury, C. F., Bishop, D. V.M., &Briscoe, J. (2002). Does impaired grammatical comprehension
provide evidence for an innate grammarmodule?Applied Psycholinguistics, 23, 247–268.
doi:10.1017.S0142716402002059
Novogrodsky, R.,Meir, N., &Michael, R. (2018).Morphosyntactic abilities of toddlerswith hear-
ing impairment and normal hearing: Evidence from a sentence-repetition task. Interna-
tional Journal of Language & Communication Disorders, 53(4), 811–824. doi:10.1111/
1460-6984.12386
Penke,M. &Wimmer, E. (2018). Deficits in comprehending wh-questions in children with hear-
ing loss – the contribution of phonological short-termmemory and syntactic complexity.
Clinical Linguistics & Phonetics, 32(3), 267–284. doi:10.1080/02699206.2017.1350883
Pisoni, D. D. & Geers, A. E. (2000). Working memory in deaf children with cochlear implants:
Correlations between digit span andmeasures of spoken language processing. Annals of
Otology, Rhinology, and Laryngology, 185, 92–93. doi:10.1177/0003489400109s1240
Power, D. J. & Quigley, S. P. (1973). Deaf children’s acquisition of the passive voice. Journal of
Speech and Hearing Research, 16, 5–11.
Qi, S. & Mitchell, R. E. (2012). Large-scale academic achievement testing of deaf and hard-of-
hearing students: Past, present, and future. Journal of Deaf Studies and Deaf Education,
17(1), 1–18. doi:10.1093/deafed/enr028
Quigley, S. P., Smith, N. L., & Wilbur, R. B. (1974). Comprehension of relativized sentences by
deaf students. Journal of Speech and Hearing Research, 17, 325–341. doi:10.1044/jshr.
1703.325
30
Quigley, S. P., Wilbur, R. B., & Montanelli, D. S. (1974). Question formation in the language of
deaf students. Journal of Speech and Hearing Research, 17, 699–713. doi:10.1044/jshr.
1704.699
Reynell, J. K. & Gruber, C. P. (1990). Reynell developmental language scales. Los Angeles, CA:
Western Psychological Services.
Ruigendijk, E. & Friedmann, N. (2017). A deficit in movement-derived sentences in German-
speaking hearing-impaired children. Frontiers in Psychology, 8, 689. doi:10.3389/fpsyg.
2017.00689
Schorr, E. A., Roth, F. P., & Fox, N. A. (2008). A comparison of the speech and language skills
of children with cochlear implants and children with normal hearing. Communication
Disorders Quarterly, 29(4), 195–210. doi:10.1177/1525740108321217
Schouwenaars, A., Finke, M., Hendriks, P., & Ruigendijk, E. (2019). Which questions do children
with cochlear implants understand? an eye-tracking study. Journal of Speech, Language,
and Hearing Research, 62, 387–409. doi:10.1044/2018_JSLHR-H-17-031
Semel, E., Wiig, E., & Secord, W. (1995). Clinical evaluation of language fundamentals-iii. San
Antonio, TX: Psychological Corporation.
Semel-Mintz, E., Wiig, E., & Secord, W. (2003). Clinical evaluation of language fundamentals
(4th ed). San Antonio, TX: Psychological Corporation.
Soleymani, Z., Mahmoodabadi, N., & Nouri, M. M. (2016). Language skills and phonological
awareness in children with cochlear implants and normal hearing. International Journal
of Pediatric Otorhinolaryngology, 83, 16–21. doi:10.1016/j.ijporl.2016.01.013
Sundström, S., Löfkvist, U., Lyxell, B., & Samuelsson, C. (2018). Phonological and grammatical
production in children with developmental language disorder and children with hear-
ing impairment. Child Language Teaching and Therapy, 34(3), 289–302. doi:10.1177/
0265659018805202
Svirsky, M. A., Stallings, L. M., Lento, C. L., Ying, E., & Leonard, L. B. (2002). Grammatical mor-
phologic development in pediatric cochlear implant users may be affected by the per-
ceptual prominence of the relevant markers. The Annals of Otology, Rhinology & Laryn-
gology. Supplement, 189, 109–112.
Szagun,G. (2000). The acquisition of grammatical and lexical structures in childrenwith cochlear
implants: A developmental psycholinguistic approach. Audiology and Neurotology, 5(1),
39–47. doi:10.1159/000013864
31
Szagun, G. (2004). Learning by ear: On the acquisition of case and gender marking by German-
speaking children with cochlear implants and with normal hearing. Journal of Child Lan-
guage, 31, 1–30. doi:10.1017/S0305000903005889
Teymouri, R., Daneshmandan, N., Hemmati, S., & Soleimani, F. (2014). Perception development
of complex syntactic construction in children with hearing impairment. Iranian Rehabil-
itation Journal, 12(22), 11–16. Retrieved from http :// irj .uswr .ac . ir/article - 1 - 426 -
en.html
Tomblin, J. B., Barker, B. A., Spencer, L. J., Zhang, X., & Gantz, B. J. (2005). The effect of age
at cochlear implant initial stimulation on expressive language growth in infants and tod-
dlers. Journal of Speech, Language, and Hearing Research, 48(4), 853–867. doi:10.1044/
1092-4388(2005/059)
Tomblin, J. B., Harrison, M., Ambrose, S. E., Walker, E. A., Oleson, J. J., & Moeller, M. P. (2015).
Language outcomes in young children with mild to severe hearing loss. Ear and Hearing,
36, 76S–91S. doi:10.1097/AUD.0000000000000219
Tuller, L. & Delage, H. (2014). Mild-to-moderate hearing loss and language impairment: How
are they linked? Lingua, 139, 80–101. doi:10.1016/j.lingua.2013.10.009
Volpato, F. (2010). The acquisition of relative clauses and phi-features in hearing and hearing-
impaired populations (Doctoral dissertation, Ca’ Foscari University of Venice). Retrieved
from http://dspace.unive.it/handle/10579/1024
Volpato, F. (2020). Verbal workingmemory resources and comprehension of relative clauses in
children with cochlear implants. First Language, 1–21. doi:10.1177/0142723719900739
Volpato, F. & D’Ortenzio, S. (2018). Ask a question! how Italian children with cochlear im-
plants produce subject and object wh-questions. Bucharest Working Papers in Linguis-
tics, 20(2), 53–76.
Volpato, F.&Vernice,M. (2014). The production of relative clauses by Italian cochlear-implanted
and hearing children. Lingua, 139, 39–67. doi:10.1016/j.lingua.2013.10.010
Werfel, K. L. (2018). Morphosyntax production of preschool children with hearing loss: An eval-
uation of the extended optional infinitive and surface accounts. Journal of Speech Lan-
guage and Hearing Research, 61(9), 2313. doi:10.1044/2018_jslhr-l-17-0406
Werfel, K. L., Reynolds, G., Hudgins, S., Castaldo, M., & Lund, E. A. (2021). The production of
complex syntax in spontaneous language by 4-year-old childrenwith hearing loss.Ameri-
32
can Journal of Speech-Language Pathology, 30(2), 609–621. doi:10.1044/2020_AJSLP-
20-00178
Wimmer, E., Rothweiler, M., & Penke, M. (2017). Acquisition of who-question comprehension
in German children with hearing loss. Journal of Communication Disorders, 67, 35–48.
doi:10.1016/j.jcomdis.2017.05.003
Woolfolk-Carrow, E. (1999). Test for auditory comprehension of language. Austin, TX: Pro-ed.
Young, G. A. & Killen, D. H. (2002). Receptive and expressive language skills of children with
five years of experience using a cochlear implant. Annals of Otology, Rhinology & Laryn-
gology, 111, 802–810. doi:10.1177/000348940211100908
33
Appe
ndixA
TableA1:List
ofstud
ieso
nthecompreh
ensio
nandprod
uctio
nof
specifictypes
ofsenten
cesb
yCH
I.Stud
iesinclude
dandexclud
edinthisreview
,
aswellasthereason
sfore
xclusio
nin
thelatte
rcase,
areindicated.
SA=Simpleactivesenten
ces.
Top=Topicalized
senten
ces.
Pass
=Passive
senten
ces.Re
l=Senten
cesw
ithrelativeclauses.Wh-Q=Wh-qu
estio
ns.O
th=Others.Incl=Inclusionde
cisio
n.HE
=He
aringeq
uipm
ent.MOC=
Mod
eof
commun
icatio
n.
Stud
yLangua
geSenten
cestructure
Incl
Reason
sfor
exclusion
SATop
Pass
Rel
Wh-Q
Oth
Friedm
annandHa
ddad-Hanna,2014
Arabic
xx
xx
yes
Power
andQuigley,1973
English
xno
Noinform
ationon
HEno
rMOC
Quigley,Smith
,and
Wilbur,1974
English
xno
Noinform
ationon
HEno
rMOC
Quigley,W
ilbur,and
Mon
tane
lli,1974
English
xno
Noinform
ationon
HEno
rMOC
Davisa
ndBlasde
ll,1975
English
xyes
Kitzingera
ndJohn
,198
0English
xx
noNo
inform
ationon
HEno
rMOC
deVilliers,1988
English
xx
xno
Noinform
ationon
HE
Norburyet
al.,20
02English
xx
yes
DeLuca,2015
English
xyes
Werfeleta
l.,20
21English
xx
yes
Delage,200
8Fren
chx
xyes
Tullera
ndDe
lage,2014
Fren
chx
yes
34
Stud
yLangua
geSenten
cestructure
Incl
Reason
sfor
exclusion
SATop
Pass
Rel
Wh-Q
Oth
Ruigen
dijkandFriedm
ann,20
17Ge
rman
xx
xx
xyes
Wim
mer
etal.,20
17Ge
rman
xyes
PenkeandWim
mer,2018
Germ
anx
yes
Scho
uwen
aarset
al.,20
19Ge
rman
xyes
Friedm
annandSzterm
an,200
6He
brew
xx
xyes
Friedm
annet
al.,20
08He
brew
xx
xyes
Friedm
annandCo
sta,20
11He
brew
,
Arabic
xyes
Friedm
annandSzterm
an,2011
Hebrew
xx
xyes
Friedm
annet
al.,20
16He
brew
xx
yes
Novogrod
skyet
al.,20
18He
brew
xx
yes
Volpato,20
10Ita
lian
xyes
Caselliet
al.,20
12Ita
lian
xyes
VolpatoandVe
rnice,20
14Ita
lian
xyes
VolpatoandD’Orten
zio,2018
Italian
xyes
D’Orten
zioandVo
lpato,20
20Ita
lian
xyes
Volpato,20
20Ita
lian
xyes
Fujiyoshiet
al.,20
12Japane
sex
xx
xx
yes
35
Stud
yLangua
geSenten
cestructure
Incl
Reason
sfor
exclusion
SATop
Pass
Rel
Wh-Q
Oth
Leeet
al.,20
18Ko
rean
xx
yes
Teym
ouri,
Dane
shmandan,He
mmati,
andSoleim
ani,20
14
Persian
xx
xx
noNo
inform
ationon
HEno
rMOC
CostaandMoita,2016
Portugue
sex
noNo
statistica
lanalysesrep
orted
Moita
andLobo
,2018
Portugue
sex
noNo
statistica
lanalysesrep
orted
Sund
ström
etal.,20
18Sw
edish
xx
xx
noNo
teno
ughinform
ationon
senten
cestructure
36
Appe
ndixB
TableB1:C
orrelatio
nbe
tweenclinica
lvariables
andthecompreh
ensio
nandprod
uctio
nof
specifictype
sofsen
tences
byCH
I.N=nu
mbe
rofC
HI
insample;HL
=he
aringloss;T
1,T2
=tim
e-po
intsinlongitu
dinalstudy;com
p=compreh
ensio
n;prod
=elicite
dprod
uctio
n;sprod=spon
tane
ous
prod
uctio
n;rep=repe
tition;top=topicalized
senten
ces;pass=passivesenten
ces;RC
=senten
cesw
ithrelativeclauses;SR
andOR=subjecta
nd
objectrelativeclauses;Sw
ho-Q
andOwho
-Q=subjectand
objectwho
-que
stions;S
which-Q
andOwhich-Q
=subjectand
objectwhich-que
stions;
ns=no
nsig
nificant;+=po
sitivecorrelation;
-=ne
gativecorrelation.
Stud
yN
Ability
tested
Structure
Age
Hearing
age
Degree
ofHL
Degree
of
aide
dHL
Hearing
device
Ageat
diagno
sis
Davisa
ndBlasde
ll,1975
23comp
SRns
ns
Friedm
annandSzterm
an,200
620
comp
ORandtopO
VSns
ns
Delage,200
8(T1)
32sprod
RCns
nsns
sprod
RCns
nsns
Delage,200
8(T2)
29prod
RCns
nsns
SR+
ns
ORwith
preverbalS
nsns
Volpato,20
1013
comp
ORwith
postverbalS
nsns
37
Stud
yN
Ability
tested
Structure
Age
Hearing
age
Degree
ofHL
Degree
of
aide
dHL
Hearing
device
Ageat
diagno
sis
RCns
which-Q
nsFriedm
annandHa
ddad-Hanna,
2014
21comp
top
ns
DeLuca,2015
16comp
who
-and
which-Q
nsns
Wim
mer
etal.,20
1721
comp
who
-Qns
nsns
who
-Q+
nsns
ns
Swho
-Qns
nsns
nsPenkeandWim
mer,2018
21comp
Owho
-Q+
nsns
ns
Scho
uwen
aarset
al.,20
1921
comp
which-Q
++
TableB2
:Con
tinuatio
nof
TableB1.H
A=he
aringaid;
CI=cochlear
implant.
Stud
yN
Ability
tested
Structure
Ageof
HA fittin
g
Ageof
CI
fittin
g
Length
ofHA use
Length
ofCI
use
Ageat
onset
of
interven
tion
Length
of
speech
therap
y
Friedm
annandSzterm
an,200
620
comp
ORandtopO
VS-
ns (n=6
)
Delage,200
8(T1)
32sprod
RCns
38
Stud
yN
Ability
tested
Structure
Ageof
HA fittin
g
Ageof
CI
fittin
g
Length
ofHA use
Length
ofCI
use
Ageat
onset
of
interven
tion
Length
of
speech
therap
y
sprod
RCns
Delage,200
8(T2)
29prod
RCns
SRns
nsns
ORwith
preverbalS
nsns
nsVo
lpato,20
1013
comp
ORwith
postverbalS
nsns
ns
SRns
ns+
VolpatoandVe
rnice,20
1413
prod
OR
nsns
ns
DeLuca,2015
16comp
who
-and
which-Q
ns
Wim
mer
etal.,20
1721
comp
who
-Qns
nsns
who
-Qns
Swho
-Qns
PenkeandWim
mer,2018
21comp
Owho
-Qns
VolpatoandD’Orten
zio,2018
13prod
who
-and
which-Q
nsns
ns
Scho
uwen
aarset
al.,20
1921
comp
Owhich-Q
nsns
Owhich-Q
+D’Orten
zioandVo
lpato,20
2010
prod
Owho
-Q+
39
TableB3
:Correlatio
nbe
tweencognitive
varia
bles
andthecompreh
ensio
nandprod
uctio
nof
specifictype
sofsentences
byCH
I.N=nu
mbe
rof
CHIinsample;HA
=he
aringaid;
CI=cochlear
implant;comp=compreh
ensio
n;prod
=elicite
dprod
uctio
n;sprod=spon
tane
ousp
rodu
ction;
rep
=repe
tition;
top=topicalized
senten
ces;pass
=passivesenten
ces;RC
=senten
cesw
ithrelativeclauses;SR
andOR=subjecta
ndob
jectrelative
clauses;Swho
-QandOwho
-Q=subjecta
ndob
jectwho
-que
stions;S
which-Q
andOwhich-Q
=subjecta
ndob
jectwhich-que
stions;STM
=short-
term
mem
ory;
FDS=forw
arddigitspan;
WM
=working
mem
ory;
NWR=no
nwordrepe
tition;
ns=no
nsig
nificant;+=po
sitivecorrelation;
-=
negativecorrelation.
Stud
yN
Ability
tested
Structure
Non
verbal
abilitie
s
STM
tasks
WM
tasks
NWR
Senten
ce
rep
FDS
Wordspan
Delage,200
8(T1)
32sprod
RC+
Delage,200
8(T2)
29sprod
RCns
prod
RCns
RCns
+ns
nsns
SRns
nsns
nsns
ORwith
preverbalS
nsns
nsns
nsVo
lpato,20
1013
comp
ORwith
postverbalS
nsns
nsns
ns
DeLuca,2015
16comp
who
-and
which-Q
nsns
Leeet
al.,20
1820
comp
pass
+
PenkeandWim
mer,2018
21comp
who
-Qns
ns
Swho
-Qns
ns
Owho
-Qns
ns
40
Stud
yN
Ability
tested
Structure
Non
verbal
abilitie
s
STM
tasks
WM
tasks
NWR
Senten
ce
rep
FDS
Wordspan
Scho
uwen
aarset
al.,20
1921
comp
Owhich-Q
+
Volpato,20
2013
comp
SRns
nsns
ORwith
preverbalS
and
NP1-NP
2nu
mbe
rmatch
nsns
ns
ORwith
preverbalS
and
NP1-NP
2nu
mbe
rmism
atch
+ns
ns
ORwith
postverbalS
+ns
ns
41
Top Related